J. Agric. Food Chem. 2005, 53, 8881−8895
8881
Optimization of the Analysis of Flavor Volatile Compounds by Liquid−Liquid Microextraction (LLME). Application to the Aroma Analysis of Melons, Peaches, Grapes, Strawberries, and Tomatoes CHRISTOPHE AUBERT,* SEVERINE BAUMANN,
AND
HELENE ARGUEL
Centre Technique Interprofessionnel des Fruits et Le´gumes, route de Molle´ge`s, 13210 Saint-Re´my-de-Provence, France
A fast method based on liquid-liquid microextraction (LLME) has been developed for the analysis of volatile compounds in fruit and vegetable juices. The method was tested in an aqueous solution containing 49 common flavor compounds typically found in fruit aroma. Influence on extraction yield of the salts used, their levels, and the time of extraction was investigated. The efficiency of n-propyl gallate to inhibit the formation of secondary compounds from lipids during the crushing of fruit tissues was also tested. The proposed method was then applied to several authentic samples such as melons, peaches, grapes, strawberries, and tomatoes. The advantages and limitations of LLME are discussed. KEYWORDS: Flavor; volatile compounds; fruits; liquid-liquid microextraction
INTRODUCTION
MATERIALS AND METHODS
Liquid-liquid extraction (LLE) is certainly one of the most frequently reported methods in the literature for the isolation of volatile compounds in fruits and vegetables. Nevertheless, because LLE requires large amounts of high-purity solvents, this technique is relatively tedious and time-consuming and causes serious environmental problems. Furthermore, extracts often have to be concentrated hundreds of times, and loss of analytes and artifact formation, due to elevated temperatures, during the concentration step are generally observed. Numerous efforts have been made to improve the LLE methods, in particular with the introduction of liquid-liquid microextraction (LLME) techniques. LLME is a single-step extraction with a very high liquid sample/solvent ratio and a saturation of the aqueous phase with inorganic salts. Microextractions avoid the concentration process and reduce solvent consumption and time of analysis. In the 1980s, Jennings (1) and Kok et al. (2) had already proposed analytical methods based on this idea, but their results were semiquantitative. Up to now, LLME has been mainly used for pesticide residue determinations in water (36) or for volatiles analysis in grape (7) and wine (8-13). Except for the method developed by Cacho et al. (7), including steam stream distillation and microextraction, no study using microextraction has appeared in the literature on the determination of aroma components in fruits and vegetables. The aim of this work was to develope a fast and quantitative method for the determination of the main volatile compounds in fruits and vegetables. Melons, peaches, grapes, strawberries, and tomatoes have been analyzed using the proposed method.
Solvent and Chemicals. Analytical grade dichloromethane (Pestanal, g99.8%), ammonium sulfate (NH4)2SO4 (puriss. p.a., g99%), ethanol absolute (Spectranal, g99.8%), and n-alkane standards (C8-C40) were from Riedel-de Hae¨n, and n-propyl gallate (g98%) was from Fluka. A stock solution containing 49 reference compounds was prepared in ethanol (Table 1). Reference compounds were obtained from various suppliers given in parentheses: no. 1-11, 13-15, 18-27, 32-37, 4042, 44, 48, and 49 (Aldrich); no. 16, 17, 30, and 38 (Fluka); no. 12, 28, 29, 31, 39, 43, and 45-47 (Interchim) (see Table 1 for compound numbers). All reference compound purities were >95%. A flavor test solution was prepared by taking 25 µL of the stock solution and diluting it to 500.0 mL with deionized water, giving ∼1 ppm of each flavor compound (except for Furaneol, ∼3 ppm). A flavor standard solution in dichloromethane was also prepared by taking 50 µL of the stock solution and diluting it to 10.0 mL. Juice Sample Preparation. Amounts of 250 g of sample (fruit or vegetable), 250 mL of n-propyl gallate (10 mM), and 25 µL of 2-octanol (3.32 µg/mL) (internal standard) were homogenized in a Waring blender for 2 min. The mixture was centrifuged (13000g, 5 min, 4 °C), and the supernatant was filtered on a stainless steel sieve (16 mesh). Optimized LLME. Forty milliliters of the flavor test solution (or sample juice previously obtained) and 12.8 g of (NH4)2SO4 (32%; w/v) were agitated until complete salt dissolution and ultracentrifuged (21000g, 5 min, 4 °C). The supernatant was then filtered through a Whatman paper filter (grade 113v) into a 50 mL screw-capped conical centrifuge tube (34 × 98 mm glass borosilicate) containing a magnetic stir bar (15 × 6 mm). Two hundred and fifty microliters of dichloromethane was added, and the mixture was extracted for 60 min under magnetic stirring at 4 °C. After removal of the magnetic stir bar, the tube was sonicated for 1 min in a Branson Ultrasonic Cleaner 5510 and centrifuged (1000g, 5 min, 4 °C). The dichloromethane extract was then recovered with a 50 µL syringe, transferred to a 100 µL vial, and immediately injected in GC-MS and GC-FID.
* Corresponding author (telephone +33 490 92 05 82; fax +33 490 92 48 87; e-mail
[email protected]).
10.1021/jf0510541 CCC: $30.25 © 2005 American Chemical Society Published on Web 10/21/2005
8882
J. Agric. Food Chem., Vol. 53, No. 23, 2005
Aubert et al.
Table 1. Recoveries of the Reference Compounds in the Flavor Test Solution by LLE and LLME Sampling recoverya (%) no.
RT (min)
RIb
mg/mLc
log Pd
compound
LLE
LLME
Fe
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49
7.0 7.7 8.3 9.0 9.2 9.7 10.8 12.1 12.7 13.5 14.5 15.5 17.1 19.0 19.8 20.5 21.6 22.3 22.6 23.6 24.0 25.9 26.7 27.2 28.6 29.6 30.6 31.1 33.4 33.7 34.1 34.6 36.0 36.5 37.5 39.2 39.5 40.6 41.7 44.5 47.5 47.9 48.6 49.3 50.0 50.9 52.2 53.1 55.2
1009 1028 1043 1062 1069 1084 1112 1141 1155 1175 1198 1220 1258 1298 1317 1340 1366 1378 1390 1409 1418 1464 1485 1497 1532 1561 1581 1597 1661 1669 1686 1691 1727 1746 1775 1824 1837 1871 1897 1994 2077 2091 2114 2144 2176 2220 2284 2330 2447
17.76 17.89 17.58 18.13 17.02 17.70 17.92 17.45 17.00 17.21 16.75 17.76 17.74 18.28 16.85 18.69 17.49 19.93 16.09 19.53 21.81 22.00 21.81 21.96 21.89 20.36 18.63 21.21 19.08 16.46 20.21 21.15 23.67 16.02 16.97 16.92 20.08 19.36 17.73 56.98 17.82 19.67 17.98 20.57 16.73 15.78 16.63 16.30 19.15
1.77 1.85 2.26 1.82 1.80 4.35 2.26 4.62 4.75 4.83 1.58 2.83 2.83 2.61 2.06 2.03 1.61 2.94 1.61 2.90 0.95 0.95 1.71 −0.05 2.97 0.62 3.44 1.44 0.60 3.33 0.44 2.08 2.60 3.56 3.47 3.47 1.08 1.57 4.29 0.82 2.57 2.73 2.57 2.26 3.06 3.06 3.55 3.55 1.55
isobutyl acetate ethyl butyrate ethyl 2-methylbutyrate butyl acetate hexanal β-pinene isoamyl acetate R-phellandrene R-terpinene limonene (E)-2-hexenal ethyl hexanoate hexyl acetate (Z)-3-hexenyl acetate 6-methyl-5-hepten-2-one hexanol (Z)-3-hexen-1-ol 2-isobutylthiazole (E)-2-hexen-1-ol 2-octanol ethyl 2-(methylthio)acetate 2-(methylthio)ethyl acetate benzaldehyde 2-(methylthio)ethanol linalool mesifurane β-cyclocitral 3-(methylthio)propyl acetate γ-hexalactone R-terpineol 3-(methyltio)propanol benzyl acetate methyl salicylate citronellol nerol geraniol benzyl alcohol 2-phenylethanol β-ionone Furaneol γ-decalactone eugenol δ-decalactone methyl anthranilate γ-undecalactone δ-undecalactone γ-dodecalactone δ-dodecalactone vanillin
66 (5) 71 (5) 65 (5) 71 (5) 67 (4) 12 (6) 70 (6) 15 (7) 18 (13) 21 (5) 89 (6) 67 (6) 69 (5) 77 (6) 79 (6) 80 (5) 78 (5) 80 (5) 81 (5) 84 (6) 79 (5) 79 (6) 75 (5) 38 (4) 84 (6) 81 (6) 74 (7) 81 (7) 79 (9) 81 (8) 54 (6) 79 (7) 76 (8) 81 (9) 94 (11) 98 (12) 75 (7) 81 (9) 70 (17) 22 (15) 88 (20) 83 (15) 78 (7) 87 (14) 86 (21) 77 (9) 71 (8) 82 (9) 92 (9)
66 (2) 70 (2) 61 (2) 73 (2) 72 (3) 2 (6) 68 (2) 4 (2) 5 (2) 5 (2) 96 (3) 53 (3) 53 (3) 75 (3) 89 (3) 48 (5) 27 (5) 89 (3) 37 (6) 90 (3) 96 (3) 97 (4) 82 (3) 3 (4) 94 (5) 65 (6) 63 (6) 98 (6) 75 (5) 81 (5) 4 (3) 100 (6) 89 (6) 79 (9) 94 (1) 91 (4) 8 (9) 30 (10) 26 (2) 1 (6) 91 (2) 93 (0) 93 (2) 86 (3) 58 (4) 72 (3) 24 (6) 50 (5) 32 (3)
0.1 0.2 3.1 0.4 6.4 442.9 1.0 223.7 81.8 532.1 3.9 35.7 41.9 0.5 11.7 137.3 409.4 9.4 243.3 3.3 33.3 28.8 5.9 1518.1 6.4 18.7 6.8 12.3 0.6 0.1 793.5 19.3 7.2 0.1 0.1 1.2 470.2 124.4 41.5 120.2 0.1 2.1 19.2 0.1 7.4 1.9 215.8 47.2 154.9
pf
*** *** *** *** ** ** * *** *** * *** ** ** *** * * *** **
*** *** ** *** **
*** ** ***
a Mean of three replicates with standard deviation in parentheses. b Linear retention index based on a series of n-hydrocarbons. c Concentration in the stock solution. Octanol/water partition coefficient estimated using Syracuse Research Corp., LogKow v. 1.6 (see Meylan, W. M.; Howard, P. H. J. Pharm. Sci. 1995, 84, 83−92). e F(0.05; 1; 4) ) 7.7; F(0.01; 1; 4) ) 21.2; F(0.001; 1; 4) ) 74.1. f *, p < 0.05; **, p < 0.01; ***, p < 0.001.
d
Conventional LLE. One hundred and fifty milliliters of the flavor test solution (or sample juice previously obtained) was extracted three times with 50 mL (3 × 15 min) of dichloromethane under magnetic stirring at 4 °C. The solvent was then removed using a small Vigreux column at 50 °C and concentrated to ∼500 µL. The extract was then immediately injected in GC-MS and GC-FID. Extraction Recoveries. The flavor test solution was extracted in triplicate using the optimized LLME and the conventional LLE methods. For both, the final volume of dichloromethane extract was precisely recovered and noted. To 50 µL of each recovered extract was added 50 µL of 4-nonanol (50 ppm), and the mixtures were homogenized and injected in GC-FID. The recoveries were calculated as a percent, based on the ratio of the peak areas of the reference compounds extracted to the peak area of 4-nonanol, using calibration graphs for each standard constructed using solutions at known concentrations. GC-FID Conditions. A Varian 3800 gas chromatograph equipped with a SolGel-Wax (SGE) capillary column (30 m × 0.25 mm i.d.,
0.25 µm film thickness) was used. The flow of hydrogen 5.7 carrier gas was 1 mL/min. The oven temperature was kept at 35 °C for 5 min, then programmed to 150 °C at 3 °C/min and to 250 °C at 5 °C/min, and kept at 250 °C for 10 min. Injections (2 µL) were performed at 220 °C in splitless mode (3 min) using a CombiPAL autosampler (CTC Analytics). The FID detector was kept at 250 °C. The levels of the volatile compounds in the authentic sample juices were expressed as 2-octanol equivalents (assuming all of the response factors were 1). The concentrations are to be considered as relative data because recovery after extraction and calibration factors related to the standard were not determined. GC-MS Conditions. A Varian 3800 gas chromatograph was used with the same SolGel-Wax capillary column as above. Injections (1 µL) were performed at 220 °C in splitless mode (3 min) using a CombiPAL autosampler (CTC Analytics). The flow of helium 6.0 carrier gas was 1 mL/min. The oven temperature program was as above. A Saturn ion-trap mass spectrometer was used. Mass spectra were
Flavor Volatiles Analysis by LLME
J. Agric. Food Chem., Vol. 53, No. 23, 2005
8883
Figure 1. GC-FID chromatograms of the flavor test solution by direct injection (A) and by LLME sampling (B) (for component listing, see Table 1). recorded in electron impact (EI) ionization mode. The ion trap, the manifold, and the transfer line temperatures were set, respectively, at 150, 45, and 250 °C. Mass spectra were scanned in the range m/z 30350 amu at 1 s intervals. Identifications were carried out by comparison of linear retention indices and EI mass spectra with data from authentic compounds. RESULTS AND DISCUSSION
LLME Performance. LLME was tested for common flavor compounds typically found in fruit aroma. Figure 1 shows the
GC-FID separation of the 49 volatile compounds from the flavor mixture (Table 1) both by direct injection and using the optimized microextraction procedure. The octanol/water partition coefficient (log P) and the linear retention index (RI) of the flavor compounds are shown in Table 1. The performance of the optimized LLME was compared to LLE by calculating the recoveries of the 49 flavor standards by both methods. A oneway analysis of variance was performed to determine the significant differences between LLME and LLE. The F test and
8884
J. Agric. Food Chem., Vol. 53, No. 23, 2005
Aubert et al.
Figure 2. Effects of different (NH4)2SO4 concentrations on the relative levels of reference compounds extracted by LLME (for component listing, see
Table 1).
the p value of the ANOVA are also given in Table 1. As seen in Figure 1 and Table 1, under the experimental conditions employed, most of the compounds are well extracted by LLME, particularly esters, thioesters, aldehydes, nitro compounds, terpenols, and lactones (except γ-dodecalactone). Generally, for these compounds, most of recoveries were found to be similar between the two methods, except for thioesters, benzyl acetate, δ-decalactone, 2-isobutylthiazole, and 6-methyl-5-hepten-2-one, for which recoveries by LLME were found to be significantly higher. In contrast, those observed for γ-hexalactone, mesifurane, hexyl acetate, ethyl hexanoate, and δ-dodecalactone were found to be significantly lower by the LLME method. As seen
in Table 1, the main significant differences observed between the two methods were observed for C6 alcohols, β-ionone, and three shikimic-derivatives (vanillin, 2-phenylethanol, and benzyl alcohol) and more particularly for terpenes, thioalcohols, and Furaneol, which are, respectively, fairly and badly extracted by the LLME method. As observed in Table 1, except for ethyl 2-(methylthio)acetate, 2-(methylthio)ethyl acetate, mesifurane, and γ-hexalactone, the efficiency of the LLME method is very poor for compounds with a log P lower or higher than 1, and it would seem that the most polar, and/or apolar, compounds are rather badly extracted by this method. Nevertheless, it is noteworthy that terpenes and Furaneol are also fairly extracted
Flavor Volatiles Analysis by LLME
J. Agric. Food Chem., Vol. 53, No. 23, 2005
8885
Figure 3. Effects of different times of extraction on the relative levels of reference compounds extracted by LLME (for component listing, see Table 1).
by the conventional LLE. Table 1 shows the relative standard deviations (%RSD) for the recovery of flavor compounds by LLME. Most of the compounds, even those badly extracted, had a very good %RSD. The average relative standard deviation of 4% was judged to be satisfactory compared to that observed for LLE (8%). As observed in Figure 1, a detection limit of the order of 1-50 ppb was estimated for linalool, benzyl acetate, eugenol, geraniol, benzaldehyde, methyl salicylate, nerol, 2-octanol, γ-decalactone, δ-decalactone, 6-methyl-5-hepten-2-one,
2-isobutyl thiazole, methyl anthranilate, citronellol, β-cyclocitral, 3-(methylthio)propyl acetate, (Z)-3-hexenyl acetate, isoamyl acetate, butyl acetate, (E)-2-hexenal, ethyl 2-(methylthio)acetate, isobutyl acetate, ethyl 2-methylbutyrate, 2-(methylthio)ethyl acetate, hexyl acetate, and ethyl butyrate. Ethyl hexanoate, γ-undecalactone, R-terpineol, hexanal, β-ionone, γ-hexalactone, hexanol, δ-dodecalactone, mesifurane, 2-phenylethanol, γ-dodecalactone, (E)-2-hexen-1-ol, δ-undecalactone, (Z)-3-hexen-1ol, vanillin, benzyl alcohol, R-terpinene, β-pinene, limonene,
8886
J. Agric. Food Chem., Vol. 53, No. 23, 2005
Aubert et al.
Figure 4. Effects of n-propyl gallate during homogenization on the levels of C6 compounds in tomato (for a given compound, the levels with different letters are significantly different at p < 0.05).
and R-phellandrene could be detected at 0.05-1 ppm. A detection limit of 1 ppm was estimated for 2-(methylthio)ethanol, 3-(methylthio)propanol, and Furaneol. Extraction Optimization. In this study, different preliminary experiments were carried out using different salts (ammonium sulfate, sodium chloride, and calcium chloride). They have shown that ammonium sulfate was the most appropriate for achieving a good salting out effect. Use of sodium and/or calcium chloride (20-80%; w/v) generally resulted in a lower organic phase recovery and a high emulsion formation tendency. Figure 2 shows the results of the extraction with different amounts of (NH4)2SO4 (12.8-22.4 g of salt for 40 mL of sample, i.e., 32-56%; w/v). Except for hexanol (16), (Z)-3hexen-1-ol (17), (E)-2-hexen-1-ol (19), 2-(methylthio)ethanol (24), benzyl alcohol (37), 2-phenylethanol (38), Furaneol (40), and vanillin (49), the FID response drastically decreased with increased amounts of salt, in particular for ethyl hexanoate (12), hexyl acetate (13), linalool (25), β-cyclocitral (27), nerol (35), geraniol (36), citronellol (34); β-ionone (39); γ-decalactone (41), γ- and δ-undecalactone (45, 46), and γ- and δ-dodecalactone (47, 48). Best results were observed with 32% ammonium sulfate. The time of extraction has also a great influence on the extraction efficiency. Figure 3 shows the results of the extraction with different times of extraction (15-60 min). Best results were observed with 60 min. Application to Real Samples. Because six-carbon (C6-) volatilessincluding the aldehydes hexanal, (E)-2-hexenal, and (Z)-3-hexenal, as well as their corresponding alcoholssin fruits and vegetables are known to be mainly produced from lipids during the crushing of plant tissues, the efficiency of n-propyl gallate (nPG), a well-known LOX inhibitor due to its electrondonating ability (14-16), was tested to reduce the formation of these secondary compounds during processing. Figure 4 shows the levels of C6 aldehydes and alcohols in a tomato sample in which enzyme deactivation was performed at different levels of nPG during homogenization. As seen in Figure 4,
except for (Z)-3-hexen-1-ol at 5 mM of nPG, the C6 compounds showed a similar pattern. Their levels, and in particular those of C6 aldehydes, decreased drastically with increased concentrations of nPG and reached their lowest amounts for 10 mM LOX inhibitor. These results were consistent with those previously reported by Todd et al. (17) and Cass et al. (18). The optimized method using a level of 10 mM of n-propyl gallate, 32% ammonium sulfate, and an extraction time of 60 min was then applied to the analysis of the volatile compounds of a red table grape (cv. Muscat de Hambourg), a Charentais cantaloupe melon (cv. Figaro), a strawberry (cv. Mara des Bois), a white peach (cv. Lauriered), and a grape tomato (cv. Cheers). Figures 5-9 show the chromatograms of the different samples obtained by both conventional LLE and LLME. The precision of the LLME method was estimated by performing three replicate extractions. The corresponding standard deviation (SD) was calculated and expressed as a percentage (%RSD). Data are reported in Tables 2-6. Charentais Melon. As seen in Figure 5, most of the flavor compounds extracted by conventional dichloromethane extraction were also extracted by LLME, albeit with somewhat different relative recoveries. The sensitivity of LLME in this example was generally comparable, even higher, than that of the conventional method for most compounds, particularly for esters (methyl 2-methylbutanoate, 2-methylpropyl acetate, ethyl butanoate, ethyl 2-methylbutanoate, butyl acetate, 2-methylbutyl acetate, ethyl hexanoate, and hexyl acetate) and thioesters [methyl 2-(methylthio)acetate, ethyl 2-(methylthio)acetate, 2-(methylthio)ethyl acetate, and 3-(methylthio)propyl acetate], which are generally recognized as the most important volatile constituents of Charentais melon aroma (19-23). In contrast, alcohols (2-methylpropanol, butanol, isoamyl alcohol, 2-phenylethanol), acetoin (3-hydroxy-2-butanone), and thioalcohols [3-(methylthio)propanol and 2-(methylthio)ethanol] were, respectively, fairly and badly extracted. As observed in Table 2, except for (Z)-3-hexenyl acetate, benzaldehyde, and 3-phenyl-
Flavor Volatiles Analysis by LLME
J. Agric. Food Chem., Vol. 53, No. 23, 2005
8887
Figure 5. GC-FID chromatograms of a Charentais cantaloupe melon (cv. Figaro) by (A) conventional dichloromethane extraction and (B) LLME [1, methyl 2-methylbutanoate; 2, 2-methylpropyl acetate; 3, ethyl butanoate; 4, ethyl 2-methylbutanoate; 5, butyl acetate; 6, 2-methylpropanol; 7, 2-methylbutyl acetate; 8, butanol; 9, 2-methylbutanol; 10, ethyl hexanoate; 11, hexyl acetate; 12, 3-hydroxy-2-butanone (acetoin); 13, (Z)-3-hexenyl acetate; 14, 6-methyl-5-hepten-2-one; 15, hexanol; 16, (Z)-3-hexen-1-ol; 17, methyl 2-(methylthio)acetate; I.S., 2-octanol; 18, ethyl 2-(methylthio)acetate; 19, acetic acid; 20, 2-(methylthio)ethyl acetate; 21, 2,3-butanediol diacetate (isomer 1); 22, benzaldehyde; 23, ethyl 3-hydroxybutanoate; 24, 2-(methylthio)ethanol; 25, 2,3-butanediol diacetate (isomer 2); 26, ethyl 3-(methylthio)propanoate; 27, 3-(methylthio)propyl acetate; 28, γ-hexalactone; 29, 3-(methylthio)propanol; 30, benzyl acetate; 31, benzyl alcohol; 32, 2-phenylethanol; 33, γ-octalactone; 34, 3-phenyl-1-propanol; 35, 3-phenyl-2-propen-1-ol; 36, dihydroactinidiolide].
8888
J. Agric. Food Chem., Vol. 53, No. 23, 2005
Aubert et al.
Figure 6. GC-MS chromatograms of a grape tomato (cv. Cheers) by (A) conventional dichloromethane extraction and (B) LLME [1, hexanal; 2, isobutyl cyanide; 3, (E)-2-pentenal; 4, (Z)-3-hexenal; 5, isoamyl alcohol; 6, (E)-2-hexenal; 7, 1-nitro-3-methylbutane; 8, 6-methyl-5-hepten-2-one; 9, hexanol; 10, (Z)-3-hexen-1-ol; 11, 2-isobutylthiazole; I.S., 2-octanol; 12, 6-methyl-5-hepten-2-ol; 13, 2-ethyl-1-hexanol; 14, benzaldehyde; 15, isovaleric acid; 16, geranial; 17, methyl salcylate; 18, guaiacol; 19, benzyl alcohol; 20, 2-phenylethanol; 21, 1-nitro-2-phenylethane; 22, eugenol; 23, dihydroactinidiolide].
2-propen-1-ol, the relative standard deviation values were below 10%. With an average relative standard deviation of 7.2%, the precision of the proposed method was judged to be satisfactory compared with other isolation procedures. Grape Tomato. As seen in Figure 6, C6 compounds [hexanal, (Z)-3-hexenal, (E)-2-hexenal, and (Z)-3-hexen-1-ol], nitro compounds (isobutyl cyanide, 1-nitro-3-methylbutane, 2-isobutylthiazole, and 1-nitro-2-phenylethane), methyl salicylate, and
2-phenylethanol, compounds previously identified as important aroma components in tomato flavor (24, 25), were generally better extracted by LLME. As previously observed, polar compounds such as isovaleric acid and isoamyl alcohol were poorly extracted compared to the classical LLE. As indicated in Table 3, the %RSD values ranged between 1 and 8.6%. With an average relative standard deviation of 5.2%, the precision of the method in this example was judged to be very good.
Flavor Volatiles Analysis by LLME
J. Agric. Food Chem., Vol. 53, No. 23, 2005
8889
Figure 7. GC-FID chromatograms of a red table grape (cv. Muscat de Hambourg) by (A) conventional dichloromethane extraction and (B) LLME [1, hexanal; 2, butanol; 3, 3-penten-2-ol; 4, (E)-2-hexenal; 5, 3-hydroxy-2-butanone; 6, 4-methyl-2-pentanol; 7, hexanol; 8, (Z)-3-hexen-1-ol; I.S., 2-octanol; 9, (E)-furan linalool oxide; 10, acetic acid; 11, (Z)-furan linalool oxide; 12, linalool; 13, 3,7-dimethyl-1,5,7-octatrien-3-ol (hotrienol); 14, 1-methyl-2pyrrolidinone; 15, R-terpineol; 16, (E)-pyran linalool oxide; 17, (Z)-pyran linalool oxide; 18, citronellol; 19, nerol; 20, hexanoic acid; 21, geraniol; 22, benzyl alcohol; 23, 2-phenylethanol; 24, 3,7-dimethyl-1,7-octadiene-3,7-diol (diol 37); 25, 2,6-dimethyl-1,7-octadiene-3,6-diol (diol 36); 26, geranic acid].
Red Table Grape. The importance of the terpenoid fraction in the characteristic “floral” aroma of Muscat grapes is now widely accepted, linalool, geraniol, nerol, citronellol, R-terpineol, and linalool oxides being, qualitatively and quantitatively, the main volatile compounds (26). As shown in Figure 7, excepted
for butanol, 3-penten-2-ol, acetoin (3-hydroxy-2-butanone), 1-methyl-2-pyrrolidinone, and benzyl alcohol, the sensitivity of the LLME is comparable to that of the conventional method, in particular for C6 aldehydes [hexanal and (E)-2-hexenal], monoterpenols (linalool, geraniol, nerol, citronellol, and R-ter-
8890
J. Agric. Food Chem., Vol. 53, No. 23, 2005
Aubert et al.
Figure 8. GC-MS chromatograms of a white peach (Lauriered) by (A) conventional dichloromethane extraction and (B) LLME [1, hexanal; 2, (E)-2hexenal; 3, (Z)-3-hexenyl acetate; 4, hexanol; 5: (Z)-3-hexen-1-ol; 6, (E)-2-hexen-1-ol; I.S., 2-octanol; 7, acetic acid; 8, benzaldehyde; 9, linalool; 10, γ-hexalactone; 11, R-terpineol; 12, γ-heptalactone; 13, γ-octalactone; 14, 3,7-dimethyl-1,5-octadiene-3,7-diol; 15, γ-nonalactone; 16, unknown C13 norisoprenoid; 17, γ-decalactone; 18, γ-jasmolactone and 6-pentyl-R-pyrone; 19, δ-decalactone; 20, (Z)-7-decen-5-olide; 21, 3-hydroxy-7,8-dihydro-βionone (isomer 1); 22, 3-hydroxy-7,8-dihydro-β-ionone (isomer 2); 23, 3-hydroxy-7,8-dihydro-β-ionol; 24, 3-hydroxy-β-ionone; 25, 3-hydroxy-5,6-epoxyβ-ionone; 26, dehydrovomifoliol].
pineol), linalool oxides [furan and pyran linalool oxides (E, Z)], and geranic acid. As mentioned in Table 4, the %RSD values, ranging between 0.6 and 11.2%, were acceptable, the average relative standard deviation being ∼6%.
White Peach. Peach and nectarine volatiles have been intensively investigated, and ∼100 compounds have been identified. Among them, lactones, particularly γ- and δ-decalactone, have been reported as character impact compounds in
Flavor Volatiles Analysis by LLME
J. Agric. Food Chem., Vol. 53, No. 23, 2005
8891
Figure 9. GC-FID chromatograms of a strawberry (cv. Mara des Bois) by (A) conventional dichloromethane extraction and (B) LLME [1, ethyl butanoate; 2, butyl acetate; 3, hexanal; 4, methyl hexanoate; 5, (E)-2-hexenal; 6, butyl acetate; 7, ethyl hexanoate; 8, hexyl acetate; 9, (E)-2-hexenyl acetate; 10, hexanol; 11, (Z)-3-hexen-1-ol; 12, (E)-2-hexen-1-ol; I.S., 2-octanol; 13, (E)-furan linalool oxide; 14, (Z)-furan linalool oxide; 15, linalool; 16, 2-methylpropanoic acid; 17, mesifurane; 18, butanoic acid; 19, 2-methylbutanoic acid; 20, benzyl acetate; 21, methyl salicylate; 22, methyl nicotinate; 23, hexanoic acid; 24, Furaneol; 25, nerolidol; 26, eugenol; 27, δ-decalactone; 28, methyl anthranilate].
peach flavor. The importance of other volatiles, such as C6 aldehydes, terpenes, and C13 norisoprenoid compounds, has been also previously described (27-31). As seen in Figure 8, most of the compounds were generally better extracted by LLME, in particular C6 aldehydes [hexanal and (E)-2-hexenal], linalool,
lactones [γ- C6, C7, C8, C10, δ-C10, and (Z)-7-decen-5-olide], and C13 norisoprenoid compounds (3-hydroxy-β-ionone, 3-hydroxy-5,6-epoxy-β-ionone, and 4,5-dihydrovomifoliol). Table 5 shows the precision of the method in this example. Except for an unknown C13 norisoprenoid compound and 3-hydroxy-
J. Agric. Food Chem., Vol. 53, No. 23, 2005
8892
Aubert et al.
Table 2. Levelsa of Volatile Compounds in Charentais Melon (Cv.
Table 3. Levelsa of Volatile Compounds in Grape Tomato (Cv.
Figaro)
Cheers)
no.
RIb
assignmentc
1 2 3 4 5 6 7 8 9 10 11 12
1003 1009 1028 1043 1062 1068 1112 1118 1182 1220 1258 1265
A A A A A A A A A A A A
13 14 15 16 17
1298 1317 1340 1366 1386
A A A A A
18
1418
A
20
1464
A
21
1483
C
22 23 24 25
1485 1497 1502 1521
A C A C
26
1561
A
27 28 29 30 31 32 33 34 35 36
1597 1661 1686 1691 1837 1871 1878 1985 2121 2156
A A A A A A A C C C
compound
mean
SD
%RSD
no.
RIb
assignmentc
compound
mean
SD
%RSD
methyl 2-methylbutanoate 2-methylpropyl acetate ethyl butanoate ethyl 2-methylbutanoate butyl acetate 2-methylpropanol 2-methylbutyl acetate butanol 2-methylbutanol ethyl hexanoate hexyl acetate 3-hydroxy-2-butanone (acetoin) (Z)-3-hexenyl acetate 6-methyl-5-hepten-2-one hexanol (Z)-3-hexen-1-ol methyl 2-(methylthio)acetate ethyl 2-(methylthio)acetate 2-(methylthio)ethyl acetate 2,3-butanediol diacetate (isomer 1) benzaldehyde ethyl 3-hydroxybutanoate 2-(methylthio)ethanol 2,3-butanediol diacetate (isomer 2) ethyl 3-(methylthio)propanoate 3-(methylthio)propyl acetate γ-hexalactone 3-(methylthio)propanol benzyl acetate benzyl alcohol 2-phenylethanol γ-octalactone 3-phenyl-1-propanol 3-phenyl-2-propen-1-ol dihydroactinidiolide
147.3 579.3 1488.1 1040.2 211.5 160.6 397.8 82.5 641.5 50.5 24.4 1330.2
10.1 43.6 92.9 86.7 13.8 11.6 30.2 5.6 33.4 4.6 2.4 101.5
6.9 7.5 6.2 8.3 6.5 7.2 7.6 6.8 5.2 9.2 9.8 7.6
3.6 5.6 415.0 891.2 4.4
0.5 0.3 20.2 34.0 0.2
14.7 5.8 4.9 3.8 3.7
331.2
15.8
4.8
71.8
3.5
4.9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
1069 1090 1096 1111 1182 1198 1308 1317 1340 1366 1378 1461 1488 1485 1581 1717 1727 1826 1837 1871 2032 2091 2156
A C C A A A B (25) A A A A C C A A A A A A A B (25) A C
hexanal isobutyl cyanide (E)-2-pentenal (Z)-3-hexenal isoamyl alcohol (E)-2-hexenal 1-nitro-3-methylbutane 6-methyl-5-hepten-2-one hexanol (Z)-3-hexen-1-ol 2-isobutylthiazole 6-methyl-5-hepten-2-ol 2-ethyl-1-hexanol benzaldehyde isovaleric acid geranial methyl salicylate guaiacol benzyl alcohol 2-phenylethanol 1-nitro-2-phenylethane eugenol dihydroactinidiolide
45.9 124.2 18.5 887.1 208.8 211.5 755.2 33.7 2.4 50.1 86.0 21.2 44.1 11.1 7.4 1.6 29.9 7.6 16.0 187.5 43.8 12.2 12.4
2.5 6.4 1.1 33.6 17.5 16.0 22.1 1.2 0.1 3.3 0.9 0.5 0.7 0.6 0.6 0.1 1.3 0.6 0.6 9.2 1.6 0.9 1.1
5.4 5.2 6.0 3.8 8.4 7.5 2.9 3.6 4.0 6.5 1.0 2.2 1.5 5.0 7.9 3.9 4.4 8.2 3.6 4.9 3.8 7.6 8.6
198.2
10.6
5.3
5.8 89.9 48.3 164.4
1.3 6.0 2.3 9.4
21.8 6.6 4.8 5.7
101.3
4.2
4.1
170.8 24.7 35.9 184.1 575.5 166.9 57.8 195.8 42.7 90.3
9.4 2.3 2.4 10.7 33.8 8.5 5.0 15.0 4.3 7.9
5.5 9.3 6.7 5.8 5.9 5.1 8.7 7.7 10.2 8.8
a Values expressed in µg/kg equiv of 2-octanol. b Linear retention index based on a series of n-hydrocarbons. c A, identity confirmed by comparing mass spectra and retention time with those of authentic standards; C, tentatively identified.
7,8-dihydro-β-ionone (isomer 2), the %RSD values, ranging between 0.6 and 12.8%, were judged to be satisfactory, the average relative standard deviation being ∼7%. Strawberry. Because of its typical aroma, the volatile compounds of strawberry have been extensively studied, and >360 volatiles have been reported (37-43). Due to their very low odor threshold and their large amounts in several cultivars, Furaneol [2,5-dimethyl-4-hydroxy-3(2H)-furanone] (DHF) and mesifurane [2,5-dimethyl-4-methoxy-3(2H)-furanone] (DMF) are considered to be the two major flavor contributors. Previous studies have shown that linalool and esters, in particular, ethyl 2-methylbutanoate, methyl and ethyl butanoates, methyl and ethyl hexanoates, and hexyl and (E)-hex-2-enyl acetates, also significantly contribute to the strawberry aroma. Methyl anthranilate, which is characterized by an intense spicy-aromatic and flowery note, could be, according to Ulrich et al. (44), responsible for the typical character of the wild strawberry aroma. Finally, acids are generally described to have no impact on strawberry flavor, except for 2-methylbutanoic acid due to its relatively low odor threshold (180 ppb). As seen in Figure
a Values expressed in µg/kg equiv of 2-octanol. b Linear retention index based on a series of n-hydrocarbons. c A, identity confirmed by comparing mass spectra and retention time with those of authentic standards; B, identity tentatively assigned by comparing mass spectra with those obtained from the literature numbered and italicized in parentheses; C, tentatively identified.
Table 4. Levelsa of Volatile Compounds in Red Table Grape (Cv. Muscat de Hambourg) no.
RIb
assignmentc
1 4 7 8 9 11 12 13 15 16 17 18 18 21 22 23 24
1069 1198 1340 1366 1430 1460 1532 1602 1668 1716 1741 1746 1775 1824 1837 1871 1910
A A A A A A A C A A A A A A A A A
25
2042
A
26
2151
A
compound
mean
SD
%RSD
hexanal (E)-2-hexenal hexanol (Z)-3-hexen-1-ol (E)-furan linalool oxide (Z)-furan linalool oxide linalool hotrienol R-terpineol (E)-pyran linalool oxide (Z)-pyran linalool oxide citronellol nerol geraniol benzyl alcohol 2-phenylethanol 3,7-dimethyl-1,7-octadiene3,7-diol 2,6-dimethyl-1,7-octadiene3,6-diol geranic acid
168.2 424.5 16.7 3.3 16.4 20.6 262.9 10.4 9.3 115.0 43.7 22.1 58.4 160.8 28.0 13.7 28.0
4.0 11.0 1.1 0.4 0.8 1.1 9.6 1.0 0.7 10.0 3.8 0.6 0.4 3.5 2.7 1.3 2.2
2.4 2.6 6.6 11.2 4.6 5.1 3.7 9.5 7.0 8.7 8.7 2.7 0.6 2.2 9.6 9.2 7.9
11.0
1.0
9.4
426.9
27.8
6.5
a Values expressed in µg/kg equiv of 2-octanol. b Linear retention index based on a series of n-hydrocarbons. c A, identity confirmed by comparing mass spectra and retention time with those of authentic standards; C, tentatively identified.
9, the sensitivity of LLME is generally comparable to, or even higher than, that of the conventional LLE method for esters (ethyl butanoate, ethyl 2-methylbutanoate, butyl acetate, and methyl hexanoate), C6 dldehydes [hexanal and (E)-2-hexenal], linalool, and mesifurane [2,5-dimethyl-4-methoxy-3(2H)-furanone]. In contrast, acids (2-methylpropanoic, butanoic, 2-methylbutanoic, and hexanoic acids) and Furaneol [2,5-dimethyl-
J. Agric. Food Chem., Vol. 53, No. 23, 2005
Flavor Volatiles Analysis by LLME Table 5. Levelsa of Volatile Compounds in White Peach (Cv. Lauriered) assignmentc
no.
RIb
1 2 3 4 5 6 8 9 10 11 12 13 14
1069 1198 1298 1340 1366 1390 1485 1532 1661 1669 1777 1878 1910
A A A A A A A A A A C A C
15 16
1973 1985
C
17 18
2075 2081
A B (32)
19 20 21
2114 2121 2269
A B (27) B (33)
22
2310
B (33)
23
2315
B (34)
24 25
2317 2339
B (35) B (33)
26
2472
B (36)
8893
Table 6. Levelsa of Volatile Compounds in Strawberry (Cv. Mara des
Bois)
compound
mean
SD
%RSD
no.
RIb
assignmentc
compound
mean
SD
%RSD
hexanal (E)-2-hexenal (Z)-3-hexenyl acetate hexanol (Z)-3-hexenol (E)-2-hexenol benzaldehyde linalool γ-hexalactone R-terpineol γ-heptalactone γ-octalactone 3,7-dimethyl-1,5-octadiene-3,7-diol γ-nonalactone unknown C13 norisoprenoidd γ-decalactone γ-jasmolactone + 6-pentylR-pyrone δ-decalactone (Z)-7-decen-5-olide 3-hydroxy-7,8-dihydro-βionone (isomer 1) 3-hydroxy-7,8-dihydro-βionone (isomer 2) 3-hydroxy-7,8-dihydroβ-ionol 3-hydroxy-β-ionone 3-hydroxy-5,6-epoxyβ-ionone 4,5-dihydrovomifoliol
1263.8 1031.5 43.3 18.2 11.7 9.9 24.7 159.0 452.1 15.4 23.2 100.2 49.4
48.0 27.4 5.5 0.4 0.8 0.7 1.1 6.1 9.0 0.5 0.9 4.6 2.1
3.8 2.7 12.8 2.4 6.9 7.5 4.4 3.8 2.0 3.4 3.9 4.5 4.2
14.1 667.0
1.2 174.7
8.6 26.2
1004.3 636.1
100.8 36.9
10.0 5.8
1186.6 206.7 111.5
118.9 14.3 1.3
10.0 6.9 1.2
13.6
4.6
33.9
17.3
1.1
6.5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
1028 1043 1062 1069 1159 1198 1120 1258 1319 1340 1366 1390 1430 1461 1532 1561 1565 1624 1665 1691 1743 1761 1826 1994 2006 2091 2114 2144
A A A A A A A A A A A A A A A A A A A A A C A A C A A A
ethyl butanoate ethyl 2-methylbutanoate butyl acetate hexanal methyl hexanoate (E)-2-hexenal ethyl hexanoate hexyl acetate (E)-2-hexenyl acetate hexanol (Z)-3-hexen-1-ol (E)-2-hexen-2-ol (E)-furan linalool oxide (Z)-furan linalool oxide linalool 2-methylpropanoic acid mesifurane butanoic acid 2-methylbutanoic acid benzyl acetate methyl salicylate methyl nicotinate hexanoic acid Furaneol nerolidol eugenol δ-decalactone methyl anthranilate
515.2 66.8 42.6 27.3 207.7 165.0 33.1 10.9 42.2 21.8 5.0 34.1 5.8 9.9 251.5 34.4 1848.8 29.8 376.8 81.6 15.9 15.1 764.4 97.2 17.9 3.2 4.5 58.8
14.6 3.0 0.4 1.9 19.1 16.2 1.6 0.7 4.1 2.4 0.8 2.8 0.6 0.3 12.9 4.6 156.9 3.7 51.5 2.5 0.6 2.1 39.8 15.5 4.3 0.0 0.3 1.5
2.8 4.5 1.0 6.8 9.2 9.8 4.9 6.8 9.8 11.1 15.8 8.1 10.5 3.4 5.1 13.5 8.5 12.3 13.7 3.1 3.9 13.6 5.2 15.9 24.1 0.3 6.2 2.5
524.1 397.9
3.0 9.5
0.6 2.4
1828.0
33.7
1.8
a Values expressed in µg/kg equiv of 2-octanol. b Linear retention index based on a series of n-hydrocarbons. c A, identity confirmed by comparing mass spectra and retention time with those of authentic standards; B, identity tentatively assigned by comparing mass spectra with those obtained from the literature numbered and italicized in parentheses; C, tentatively identified. d Major mass spectral fragments [m/e (%)]: 108 (100), 43 (70), 206 (69), 93 (67), 39 (62), 150 (44), 77 (30), 41 (28), 121 (27), 136 (26).
4-hydroxy-3(2H)-furanone] are poorly extracted by LLME. Due to its water-soluble nature and its thermal instability, Furaneol (DHF) has always proved to be difficult to isolate by other isolation procedures (dynamic headspace, simultaneous distillation-extraction, or solid phase microextraction), and in most studies this compound is rarely quantified (45-53). Probably that the most reliable and accurate method for the quantitative analysis of DHF, and its methyl ether, is the stable isotope dilution analysis described by Sen et al. (54). Except for 2-methylbutanoic acid, which could contribute to strawberry aroma, the low affinity of LLME for fatty acids could be advantageous, for example, by reducing possible coelution of these compounds with other trace flavor constituents. As observed in Table 6, the %RSD values ranged between 0.3 and 24.1%. With an average relative standard deviation of 8.3%, the precision of the method in this example was judged to be acceptable. Conclusion. Traditional LLE of volatile compounds is a widely used technique. However, this method involves large amounts of solvents, laborious and time-consuming extraction phases, and a possible thermal degradation of volatiles during the concentration step. In comparison, LLME provides many advantages. LLME is fast and simple and uses microquantities of solvent. Because this method requires no concentration step,
a Values expressed in µg/kg equiv of 2-octanol. b Linear retention index based on a series of n-hydrocarbons. c A, identity confirmed by comparing mass spectra and retention time with those of authentic standards; C, tentatively identified.
LLME is less time-consuming than LLE and minimizes the risk of a thermal degradation of volatile compounds and the possible artifacts formation. This method also produces cleaner chromatograms compared with conventional LLE and prevents faster column degradation and rapid contamination of the MS source. Compared to conventional solvent extraction, except for some compounds, the sensitivity of LLME is generally comparable to, or even higher than, that for most volatile compounds. Finally, this method is generally as precise as classical extractions, or even better, and LLME is an interesting alternative for the analysis of volatile compounds in fruit and vegetable juices. LITERATURE CITED (1) Jenning, W. Recent development in high-resolution GC. In FlaVour ’81; Schreier, P., Ed.; de Gruyter: Berlin, Germany, 1981; pp 233-251. (2) Kok, M. F.; Yong, F. M.; Lim, G. Rapid extraction method for reproducible analysis of aroma volatiles. J. Agric. Food Chem. 1987, 35, 779-781. (3) Jeannot, M. A.; Cantwell, F. F. Solvent microextraction into a single drop. Anal. Chem. 1996, 68, 2236-2240. (4) Beltran, J.; Lopez, F. J.; Forcada, M.; Hernandez, F. Microextraction procedures combined with large volume injection in capillary gas chromatography for the determination of pesticide residues in environmental aqueous samples. Anal. Chim. Acta 1997, 356, 125-133. (5) Ma, M.; Cantwell, F. F. Solvent microextraction with simultaneous back-extraction for sample cleanup and preconcentration: quantitative extraction. Anal. Chem. 1998, 70, 3912-3919. (6) Fernandez-Gutierrez, A.; Martinez-Vidal, J. L.; Arrebola-Liebanas, F. J.; Gonzalez-Casado, A.; Vilchez, J. L. Determination
8894
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15) (16)
(17)
(18)
(19)
(20)
(21)
(22)
J. Agric. Food Chem., Vol. 53, No. 23, 2005 of endosulfan and some pyrethroids in waters by micro liquidliquid extraction and GC-MS. Fresenius’ J. Anal. Chem. 1998, 360, 568-572. Cacho, J.; Melendez, J.; Ferreira, V. Development of a method for analyzing volatiles from foodstuff matrices, including microextraction by demixture. Application to the analysis of grapes. Mikrochim. Acta 1992, 108, 61-72. Ferreira, V.; Rapp, A.; Cacho, J. F.; Hastrich, H.; Yavas, I. Fast and quantitative determination of wine flavor compounds using microextraction with Freon 113. J. Agric. Food Chem. 1993, 41, 1413-1420. Moio, L.; Chambellant, E.; Lesschaeve, I.; Issanchou, S.; Schlich, P.; Etievant, P. X. Production of representative wine extracts for chemical and olfactory analysis. Ital. J. Food Sci. 1995, 3, 265-277. Ferreira, V.; Escudero, A.; Lopez, R.; Cacho, J. F. Analytical characterization of the flavor of oxygen-spoiled wines through the gas chromatography-ion-trap mass spectrometry of ultratrace odorants: optimisation of conditions. J. Chromatogr. Sci. 1998, 36, 331-339. Ferreira, V.; Lopez, R.; Escudero, A.; Cacho, J. F. Quantitative determination of trace and ultratrace flavour active compounds in red wines through gas chromatographic-ion trap mass spectrometric analysis of microextracts. J. Chromatogr. A 1998, 806, 349-354. Saenz-Barrio, C.; Cedron-Fernandez, T. Microextraction of volatile compounds from wine samples and their determination by GC-FID. The effect of the salts and extraction solvents used. Chromatographia 2000, 51, 221-225. Ortega, C.; Lopez, R.; Cacho, J.; Ferreira, V. Fast analysis of important wine volatile compounds. Development and validation of a new method based on gas chromatographic-flame ionisation detection analysis of dichloromethane microextracts. J. Chromatogr. A 2001, 923, 205-214. Gunckel, S.; Santander, P.; Cordano, G.; Ferreira, J.; Munoz, S.; Nunez-Vegara, L. J.; Squella, J. A. Antioxidant activity of gallates: An electrochemical study in aqueous media. Chem. Biol. Interact. 1998, 114, 45-49. Kristie, D.; Thompson, J. E. Inhibition of lipoxygenase activity: a cautionary note. Phytochemistry 1989, 28, 2577-2581. Luning, P. A.; Carey, A. T.; Roozen, J. P.; Wichers, H. J. Characterisation and occurrence of lipoxygenase in bell peppers at different ripening stages in relation to the formation of volatile flavor compounds. J. Agric. Food Chem. 1995, 42, 1492-1500. Todd, J. F.; Paliyath, G.; Thompson, J. E. Characteristics of a membrane-associated lipoxygenase in tomato fruit. Plant Physiol. 1990, 94, 1225-1232. Cass, B. J.; Schade, F.; Robinson, C. W.; Thompson, J. E.; Legge, R. L. Production of tomato flavor volatiles from a crude enzyme preparation using a hollow-fiber reactor. Biotechnol. Bioeng. 2000, 67, 372-377. Homatidou, V.; Karvouni, S.; Dourtoglou, V. Determination of characteristic aroma components of “Cantaloupe” Cucumis melo using multidimensional gas chromatography (MDGC). In FlaVors and Off-FlaVors; Proceedings of the 6th International Flavor Conference, Crete, Greece; Charlambous, G., Ed.; Elsevier: Amsterdam, The Netherlands, 1989; pp 1011-1023. Homatidou, V. I.; Karvouni, S. S.; Dourtoglou, V. G.; Poulos, C. N. Determination of total volatile components of Cucumis melo L. variety Cantaloupensis. J. Agric. Food Chem. 1992, 40, 1385-1388. Bauchot, A. D.; Motram, D. S.; Dodson, A. T.; John, P. Effect of aminocyclopropane-1-carboxylic acid oxidase antisense gene on the formation of volatile esters in Cantaloupe Charentais melon (cv. Vedrantais). J. Agric. Food Chem. 1998, 46, 47874792. Flores, F.; El Yahyaoui, F.; de Billerbeck, G.; Romojaro, F.; Latche´, A.; Bouzayen, M.; Pech, J. C.; Ambid, C. Role of ethylene in the biosynthetic pathway of aliphatic ester aroma volatiles in Charentais Cantaloupe melons. J. Exp. Bot. 2002, 53, 201-206.
Aubert et al. (23) Aubert, C.; Bourger, N. Investigation of volatiles in Charentais Cantaloupe melons (Cucumis melo var. cantalupensis). Characterization of aroma constituents in some cultivars. J. Agric. Food Chem. 2004, 52, 4522-4528. (24) Buttery, R. G.; Teranishi, R.; Ling, L. C.; Flath, R. A.; Stern, D. J. Quantitative studies on origins of fresh tomato aroma volatiles. J. Agric. Food Chem. 1988, 36, 1247-1250. (25) Buttery, R. G.; Teranishi, R.; Flath, R. A.; Ling, L. C. Fresh tomato volatiles. Composition and sensory studies. In FlaVors Chemistry, Trends and DeVelopments; Teranishi, R., Buttery, R. G., Shahidi, F., Eds.; ACS Symposium Series 388; American Chemical Society: Washington, DC, 1989; pp 213-222. (26) Mateo, J. J.; Jimenez, M. Monoterpenes in grape juice and wines (review). J. Chromatogr. A 2000, 881, 557-567. (27) Engel, K. H.; Flath, R. A.; Buttery, R. G.; Mon, T. R.; Ramming, D. W.; Teranashi, R. Investigation of volatile constituents in nectarines. 1. Analytical and sensory characterization of aroma components in some nectarine cultivars. J. Agric. Food Chem. 1988, 36, 549-533. (28) Horvat, R. J.; Chapman, G. W.; Robertson, J. A.; Meredith, F. I.; Scorza, R.; Callahan, A. M.; Morgens, P. Comparison of the volatile compounds from several commercial peach cultivars. J. Agric. Food Chem. 1990, 38, 234-237. (29) Berger, R. G. Fruits, I. In Volatile Compounds in Foods and BeVerages; Maarse, H., Ed.; Dekker: New York, 1991; pp 291304. (30) Takeoka, G. R.; Flath, R. A.; Buttery, R. G.; Winterhalter, P.; Guntert, M.; Ramming, D. W.; Teranishi, R. Free and bound flavor constituents of white-fleshed nectarines. In FlaVor PrecursorssThermal and Enzymatic ConVersions; ACS Symposium Series 490; Teranishi, R., Takeoka, G. R., Gu¨ntert, M., Eds.; American Chemical Society: Washington, DC, 1992; pp 116-138. (31) Aubert, C.; Gu¨nata, Z.; Ambid, C.; Baumes, R. Changes in physicochemical characteristics and volatile constituents of yellow- and white-fleshed nectarines during maturation and artificial ripening. J. Agric. Food Chem. 2003, 51, 3083-3091. (32) Derail, C.; Hofmann, T.; Schieberle, P. Differences in key odorants of handmade juice of yellow-flesh peaches (Prunus persica L.) induced by the workup procedure. J. Agric Food Chem. 1999, 47, 4742-4745. (33) Krammer, G.; Winterhalter, P.; Schwab, M.; Schreier, P. Glycosidally bound aroma compounds in the fruits of Prunus species apricot (P. armenica, L.), peach (P. persica, L.), yellow plum (P. domestica, L. srp Syriaca). J. Agric. Food Chem. 1991, 39, 778-781. (34) Winterhalter, P.; Schreier, P. Free and bound C13 norisoprenoids in quince (Cydonia oblonga Mill.) fruit. J. Agric. Food Chem. 1988, 36, 1251-1256. (35) Gu¨ldner, A. P.; Winterhalter, P. Structure of two ionone glycosides from quince fruit (Cydonia oblonga Mill.). J. Agric. Food Chem. 1991, 39, 2142-2146. (36) Sefton, M.; Winterhalter, P.; Williams, P. J. Free and bound 6,9dihydroxy megastigm-7-en-3-one in Vitis Vinifera grapes and wine. Phytochemistry 1992, 31, 1813-1815. (37) Douillard, C.; Guichard, E. Comparison by multidimensional analysis of concentrations of volatile compounds in fourteen frozen strawberry varieties. Sci. Aliments 1989, 9, 53-76. (38) Latrasse, A. Fruits III. In Volatile Compounds in Foods and BeVerages; Maarse, H., Ed.; Dekker: New York, 1991; pp 329387. (39) Zabetakis, I.; Holden, M. A. A study of strawberry flavour biosynthesis. In BioflaVour ’95; Etievant, P., Schreier, P., Eds.; INRA Editions: Dijon, France, 1995; pp 211-216. (40) Schieberle, P.; Hofmann, T. Evaluation of the character impact odorants in fresh strawberry juice by quantitative measurements and sensory studies on model mixtures. J. Agric. Food Chem. 1997, 45, 227-232. (41) Zabetakis, I.; Holden, M. A. Strawberry flavour: analysis and biosynthesis. J. Sci. Food Agric. 1997, 74, 421-434.
J. Agric. Food Chem., Vol. 53, No. 23, 2005
Flavor Volatiles Analysis by LLME (42) Zabetakis, I.; Gramshaw, J. W.; Robinson, D. D. 2,5-Dimethyl4-hydroxy-2H-furan-3-one and its derivatives: analysis, synthesis and biosynthesissa review. Food Chem. 1999, 65, 139-151. (43) Me´nager, I.; Jost, M.; Aubert, C. Changes in physicochemical characteristics and volatile constituents of strawberry (cv. Cigaline) during maturation. J. Agric. Food Chem. 2004, 52, 12481254. (44) Ulrich, D.; Hoberg, E.; Rapp, A.; Kecke, S. Analysis of strawberry flavoursdiscrimination of aroma types by quantification of volatile compounds. Z. Lebensm. Unters. Forsch. 1997, 205, 218-223. (45) Pickenhagen, W.; Veluz, A.; Passerat, J. P.; Ohloff, G. Estimation of 2,5-dimethyl-4-hydroxy-3(2H)-furanone (Furaneol) in cultivated and wild strawberries, pineapples and mangoes. J. Sci. Food Agric. 1981, 32, 1132-1134. (46) Dirinck, P.; De Pooter, H.; Willaert, G.; Schamp, N. Flavor quality of cultivates strawberries: the role of the sulfur compounds. J. Agric. Food Chem. 1981, 29, 316-321. (47) Pe´rez, A. G.; Rios, J. J.; Sanz, C.; Olias, J. M., Aroma components and free amino acids in strawberry variety Chandler during ripening. J. Agric. Food Chem. 1992, 40, 2232-2235. (48) Miszczak, A.; Forney, C. F.; Prange, R. K. Development of aroma volatiles and color during postharvest ripening of ‘Kent’ strawberries. J. Am. Soc. Hortic. Sci. 1995, 120, 650-655. (49) Ulrich, D.; Eunert, S.; Hoberg, E.; Rapp, A. Strawberry aroma analysis by solid-phase microextraction. Dtsch. Lebensm.-Rundsch. 1995, 91, 349-351.
8895
(50) Ulrich, D.; Krumbein, A.; Rapp, A. Gas chromatographic analysis of aroma of strawberry, cherry, and tomato by solid-phase microextraction. Dtsch. Lebensm.-Rundsch. 1997, 93, 3311-316. (51) Forney, C. F.; Kalt, W.; McDonald, J. E.; Jordan, M. A. Changes in strawberry fruit quality during ripening on and off the plant. In Proceedings of the International PostharVest Science Conference, PostharVest ’96; Acta Hort 464; Bieleski, R., Laing, W., Clark, C., Eds.; ISHS: Leuven, Belgium, 1998; pp 506-506. (52) Song, J.; Fan, L.; Beaudry, R. M. Application of solid-phase microextraction and gas chromatography/time-of-flight mass spectrometry for rapid analysis of flavor volatiles in tomato and strawberry fruits. J. Agric. Food Chem. 1998, 46, 3721-3726. (53) Holt, R. U. Development of a method for analysing volatile flavour components in strawberry using solid-phase micro extraction and gas chromatography. In Proceedings of the 4th International Strawberry Symposium; Acta Hort 567; Hietaranta, T., Linna, M.-M., Palonen, P., Parikka, P., Eds.; ISHS: Leuven, Belgium, 2002; pp 743-746. (54) Sen, A.; Schieberle, P.; Grosch, W. Quantitative determination of 2,5-dimethyl-4-hydroxy-3(2H)-furanone and its methyl ether using a stable dilution assay. Lebensm. Wiss. Technol. 1991, 24, 365-369. Received for review May 9, 2005. Revised manuscript received July 21, 2005. Accepted September 7, 2005.
JF0510541